Systems Research Engineer/scientist

Intel Intel · Semiconductors · Oregon, Hillsboro, United States

Systems Research Engineer/Scientist role focused on leveraging AI/ML for higher efficiency and performance in system architecture innovations, including high-performance cluster computing, virtualization, and accelerated computing. The role involves prototyping, characterizing, and analyzing workloads, developing tools for performance assessment, and influencing future product roadmaps. Requires strong systems knowledge and hands-on experience with AI workloads, with a focus on performance modeling and analysis of AI inference or training.

What you'd actually do

  1. prototype, characterize, and analyze target workloads
  2. develop tools and methodology to assess the performance of future Intel products
  3. partner with others to identify and develop system enhancements to improve future Intel products
  4. Your work will influence Intel's next-generation product roadmap and architectural decisions.

Skills

Required

  • Bachelors in Computer Science, Electrical Engineering & Computer Science, Electrical & Computer Engineering, or related computing discipline with 4+ years of related experience or MS in Computer Science, Electrical Engineering & Computer Science, Electrical & Computer Engineering, or related computing discipline with 2+ years of related experience.
  • Experience in computer architecture and Intel platform architecture
  • Knowledge of and hands-on experience with system software (e.g., OS/VMM memory management, scheduling, kernel/driver architectures)
  • Experience with scripting languages such as Python/Perl/PowerShell/shell and programming languages such as C/C++
  • Experience with Linux installation/configuration
  • Experience in performance benchmarking

Nice to have

  • PhD in Computer Science, Electrical Engineering & Computer Science, Electrical & Computer Engineering, or related computing discipline
  • Experience configuring/using VMMs/containers and/or corresponding orchestration tools
  • Experience in modeling and analysis (including AI inference or training)
  • Experience with hardware and software performance monitoring resources (e.g., EMON, VTune, Perf, sysstat)
  • Experience with AI/ML frameworks and libraries (e.g., Pytorch, CUDA, vLLM, Triton, NCCL, oneCCL, oneAPI)
  • Experience with GPU performance profiling (e.g., Unitrace, Nsight, Omnitrace)
  • Experience with high-performance and cloud-native architectures and technologies (MPI, DPDK/SPDK, micro-services, containers, protobuf, gRPC)
  • Background in data science methods and tools (e.g., Pandas, R, MATLAB)
  • Experience with automation frameworks and tools

What the JD emphasized

  • AI workloads
  • AI inference or training
  • AI/ML frameworks and libraries

Other signals

  • AI/ML for higher efficiency and performance
  • prototype, characterize, and analyze target workloads
  • develop tools and methodology to assess performance
  • influence Intel's next-generation product roadmap